Introduction to ATM

1
2003-07-15
IEEE 802.20-03/65
2
Market Requirementsfor IEEE 802.20

• July 2003
• James F. Mollenauer
• Technical Strategy Associates
• Joseph R. Cleveland
• Samsung Telecommunications America
• Paul Odlyzko
• Motorola Inc.

3
The Most Effective Requirements...

• .Are simply stated
• Emperor Nero, to his banquet entertainers
• Astonish me
• President Kennedy, to NASA
• A man on the moon by the end of the decade

4
Goal for 802.20

• To be the technology of choice for mobile data
• to achieve this, we must
• Go beyond what 3G or 3G enhanced can do
• Higher in data rate
• Better in dealing with high-speed mobility
• Simpler in interfacing with IP land-line networks
• Faster in handoff
• Also, be better than 802.16e
• Otherwise, why bother?

5
The Classic Mistake

• Assuming that our next-generation system will be
  better than the competition
• Just because its better than their last one
• They are working on their next generation, too
• To succeed, we need to be have higher throughput
  than existing systems
• And higher than their next (enhanced) version

Consider ATM vs. Ethernet 25 155 Mbps vs. 10
Mbps-- Ethernet moved to 100 Mbps and wiped out
premises ATM
6
The Current State of the Art

• We need to beat these numbers by a significant
  margin

7
What is our Sustainable Technical Edge?

• Packets?
• Most new systems use packets
• More complex modulation ?
• All current systems have discovered the laws of
  physics
• CDMA?
• 2G, 2.5G, and 3G have CDMA
• OFDM?
• 802.16a has that
• Narrow channels to fit existing allocations?
• New spectrum is unlikely to require narrow
  channels
• Maybe its complex antennas
• 802.16 and even 802.11 are thinking about smart
  antennas
• Maybe its our simple IP-based protocol stack
• Nothing prevents 802.16 from using the same stack
• Bottom line if there is no single killer
  advantage, then we have to tune our system
  better. The devil is in the details.

8
The Tools We Have

• Simple protocol stack
• Conformance to normal data practice
• Voice and video support via QoS scheduling
• Smart antennas
• Great improvements in rate are possible
• But the competition can add this to existing
  standards
• Targeted primarily at mobile users
• Not an add-on
• Power consumption considered from the start
• Consistency across different channel sizes
• Easy roaming for the user and system migration
  for the operator
• Adaptive modulation and coding
• Optimize for varying signal strength and
  different multipath environments
• Smart scheduling